Re: Auto-indexing

From: Alfred Perlstein <bright(at)wintelcom(dot)net>
To: Christopher Kings-Lynne <chriskl(at)familyhealth(dot)com(dot)au>
Cc: Hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Auto-indexing
Date: 2001-02-07 02:36:16
Message-ID: 20010206183616.J26076@fw.wintelcom.net
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

* Christopher Kings-Lynne <chriskl(at)familyhealth(dot)com(dot)au> [010206 18:29] wrote:
> Is it a feasible idea that PostgreSQL could detect when an index would be
> handy, and create it itself, or at least log that a table is being queried
> but the indices are not appropriate?
>
> I suggest this as it's a feature of most windows databases, and MySQL does
> it. I think it would be a great timesaver as we have hundreds of different
> queries, and it's a real pain to have to EXPLAIN them all, etc. Is that
> possible? Feasible?

Probably both, but if it's done there should be options to:

.) disable it completely or by table/database or even threshold or
disk free parameters (indicies can be large)
.) log any auto-created databases to inform the DBA.
.) if disabled optionally log when it would have created an index on
the fly. (suggest an index)
.) expire old and unused auto-created indecies.

Generally Postgresql assumes the user knows what he's doing, but
it couldn't hurt too much to provide an option to have it assist
the user.

--
-Alfred Perlstein - [bright(at)wintelcom(dot)net|alfred(at)freebsd(dot)org]
"I have the heart of a child; I keep it in a jar on my desk."

In response to

  • Auto-indexing at 2001-02-07 02:10:24 from Christopher Kings-Lynne

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Bruce Momjian 2001-02-07 03:14:17 Re: Auto-indexing
Previous Message Christopher Kings-Lynne 2001-02-07 02:10:24 Auto-indexing